Prompt
You are a program evaluator and report writer for a small nonprofit. Help me
draft a grant report for a funder.

The grant's stated objectives and targets (from our grant agreement): {{grant_objectives}}

Our actual data for the reporting period (pasted from spreadsheets, may be
messy): {{program_data}}

Reporting period and funder: {{reporting_details}}

Do this in two parts.

PART 1 — ANALYSIS: For each grant objective, compare our actual results to the
target. State plainly: met, exceeded, or missed, with the numbers. Note any
trends across the period (growth, drop-offs, seasonal patterns) that the data
actually supports.

PART 2 — DRAFT REPORT: Write narrative sections mapped one-to-one to the grant
objectives. Where we missed a target, use honest framing: what happened, what we
learned, what we're changing — no spin, no excuses. Include a "looking ahead"
paragraph grounded only in plans I've stated.

Hard rules: use only the data I provided — never estimate, extrapolate, or fill
gaps with plausible numbers. Where data is missing for an objective, write
[DATA NEEDED: description]. If my numbers appear inconsistent, list the
discrepancies before writing anything.

Fill in your details and the prompt updates live — then copy.

What you get back (excerpt)

**Objective 2 — Distribute 250,000 lbs of food: EXCEEDED.** Actual distribution reached 268,400 lbs, 107% of target, with quarterly volume growing steadily (Q1 to Q3). **Objective 3 — Establish 3 new pantry partnerships: MISSED (2 of 3).** Grace Church and Northside Elementary partnerships launched in February and May. A third prospective partner paused talks over storage capacity. We are addressing this by [DATA NEEDED: current status of third-partner pipeline and revised timeline]. Note: unduplicated households (389) sits just below the 400 target with one quarter remaining.

The full workflow

  1. Clean obvious errors from your spreadsheet before pasting; strip client names and identifying details first.
  2. Run Part 1 and confirm every comparison against your own records before drafting narrative.
  3. Fill each [DATA NEEDED] flag from real records — or tell the funder the data isn't collected yet.
  4. Add one concrete story (with consent) that the numbers can't show; that's the part only you can write.
  5. Keep the final report and underlying data together in your grant file for the audit trail.

Watch out for

Never let AI fill data gaps with plausible-sounding numbers. Misstating results to a funder can end the relationship, and false statements on federal grant reports create False Claims Act liability.

Client and beneficiary records are confidential — remove names, addresses, and case details before pasting service data into any AI tool, and use aggregate numbers wherever possible.

Honest missed-target framing usually strengthens funder trust; AI's instinct to spin is a bug, not a feature.

Where this comes from

Every use case on this site is grounded in real reports from working nonprofit directors — not invented by us.

More AI use cases for nonprofit directors

← All 6 use cases: How Nonprofit Directors Use AI